How, exactly? I'm afraid I don't see the connection between your claim and what I said. How does addressing the subject of Wikipedia's death before addressing the subject of editors leaving mean he thinks he is "more important than his audience"? Is his audience desperate to know more about the editors, and openly contemptuous of the importance of Wikipedia's supposed decline and impending demise? Are editors his only audience, and thus he is ignoring them by addressing the death of Wikipedia first?

A meaningless statistic, taken out of context, means very little to anybody who takes the time to think about it. Broad public claims that Wikipedia is dying can do Wikipedia considerable damage, especially since it exists because of donations. He'd have to be incompetent at his job, and probably an idiot, to consider the statistic the more important of the two.

He didn't say editors are meaningless. He said that a statement about the number of editors leaving is meaningless without context. Which is true.

If you think that he is "listing half a dozen facts that have nothing to do with the question", then you don't understand the question. The question is not, "Are editors leaving Wikipedia in droves?" The question *he* cares about is, "Does this claim that editors are leaving Wikipedia in droves mean Wikipedia is dying?" So, he states outright that this claim is being made, and then disputes it. Only after he takes care of the important stuff does he address the question of number of editors. So, no, this wasn't "smoke and mirrors" (to give the actual name to the trick). It was his attempt to address what *he* considered important before talking about the accuracy of what *you* consider "the question".

You seem to be under the impression that observability creates non-randomness. It doesn't. It only creates non-randomness *if* observations done *before* the randomizing process can predict the results.

Suppose I had a device which used radioactive decay to produce perfect random whole numbers between 1 and any arbitrary number up to 52. I set an entire deck of cards in front of me, laid out side by side. I ask the device to pick a number from 1 to 52, take that card, and turn it over to one side. I then do the same again using a number between 1 and 51, placing the card face down on the first card. I continue, repeating until I have 52 perfectly chosen random cards stacked up next to me.

The deck is random, beyond any doubt, but anybody watching me knows exactly what order the deck was in. This ability to determine the final results by watching the randomizing process doesn't change the fact that it is a random result.

You can watch me shuffle from here to doomsday, but if your observations *before I begin* cannot tell you anything about the final result, this means nothing as to whether or not it is random. You have to predict the result, not observe it.

Nope. Not even a "perfect observer" could do what you describe, if the die roll is done correctly with sufficiently well-formed dice.

Quantum fluctuations influence the firing of my neurons in my brain and nerves, the twitching of my muscles, the elasticity of the dice impacts, and the movement of the molecules of the air. Further, a "perfect observer" who actually observed these quantum fluctuations would, according to quantum dynamics, inherently influence them, generating new randomness.

Further, the dice rolling is chaotic, with high sensitivity to initial conditions. Thus, even the tiniest random fluctuations in the base conditions produce completely different results. Since there is so much true unpredictability in the initial conditions, and so few possible results (only six per die), these tiny unpredictable factors translate to unpredictable macro effects.

In other words, even your "perfect observer" would still get a random result, if the die roll is rolled in such a way as to have both sufficient random fluctuations in the initial conditions and sufficient events that are sufficiently sensitive to initial conditions. However, it is theoretically possible for the perfect observer to spot when these factors do not apply, such as when a skilled dice thrower is throwing to minimize the randomizing effects.

Nope. Observed phenomena in quantum mechanics exist that could not exist if the weirder claims of quantum mechanics were not true, including the inherent perfect unpredictability (i.e. randomness ) of certain phenomena. In other words, if atomic decay could be even theoretically predicted, certain experiments that have been done would have had different results.

You're doing it wrong. What I've found works is to extrapolate to ten, one hundred, ten thousand, or a million doors, and describe it as Monty Hall going down the row of doors, looking behind them, then opening them, except one. I've never had anybody argue the point with the million doors.